In [4]:
ox.plot_graph(
leicester_osmnx_graph_prj,
node_size=3,
figsize=(10, 10))
Out[4]:
(<Figure size 1000x1000 with 1 Axes>, <Axes: >)
# Base libraries
import pandas as pd
import matplotlib.pyplot as plt
import plotly.express as px
import plotly.graph_objects as go
# NetworkX
import networkx as nx
import osmnx as ox
# OS environment setup
from local_directories import *
# Reset random seeds
random_seed = 2674
# Other
neighbourhood_min_nodes = 8
max_distance = 500
# Load Leciester's graph
leicester_osmnx_graph = ox.io.load_graphml(bulk_storage_directory + "/osmnx/raw_excluded/leicester-1864.graphml")
leicester_osmnx_graph_prj = ox.project_graph(leicester_osmnx_graph)
ox.plot_graph(
leicester_osmnx_graph_prj,
node_size=3,
figsize=(10, 10))
(<Figure size 1000x1000 with 1 Axes>, <Axes: >)
# Convert graph to dataframe version
leicester_osmnx_graph_prj_df = None
for node in leicester_osmnx_graph_prj:
node_dict = leicester_osmnx_graph_prj.nodes[node]
node_dict["osmnx_node_id"] = int(node)
# node_dict["osmnx_node_id"] = str(node)
if leicester_osmnx_graph_prj_df is None:
leicester_osmnx_graph_prj_df = pd.DataFrame.from_dict([node_dict])
else:
leicester_osmnx_graph_prj_df = pd.concat([leicester_osmnx_graph_prj_df, pd.DataFrame.from_dict([node_dict])])
leicester_osmnx_graph_prj_df.head()
| y | x | street_count | elevation | elevation_aster | elevation_srtm | lon | lat | osmnx_node_id | ref | highway | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 5.829804e+06 | 622151.977595 | 3 | 72.0 | 35 | 72 | -1.196195 | 52.604506 | 194739 | NaN | NaN |
| 0 | 5.829991e+06 | 622098.041002 | 3 | 72.0 | 45 | 72 | -1.196922 | 52.606196 | 1551014281 | NaN | NaN |
| 0 | 5.828827e+06 | 622259.813792 | 2 | 79.0 | 57 | 79 | -1.194965 | 52.595696 | 326312 | 21 | motorway_junction |
| 0 | 5.830107e+06 | 622077.742140 | 3 | 79.0 | 43 | 79 | -1.197179 | 52.607245 | 326320 | 21 | motorway_junction |
| 0 | 5.829673e+06 | 622220.645785 | 3 | 74.0 | 35 | 74 | -1.195230 | 52.603314 | 2627867454 | NaN | NaN |
# Load Leciester's base stats
leicester_osmnx_basic_stats = pd.read_csv(this_repo_directory + "/data/leicester-1864_basic_stats_dist500.csv")
leicester_osmnx_basic_stats = leicester_osmnx_basic_stats.rename(columns={"node_id":"osmnx_node_id"})
# Drop NAs created when ego-graph has less than 8 nodes
leicester_osmnx_basic_stats = leicester_osmnx_basic_stats.dropna(subset=["osmnx_node_id"])
leicester_osmnx_basic_stats.head()
| 0 | n | m | k_avg | edge_length_total | edge_length_avg | streets_per_node_avg | streets_per_node_counts | streets_per_node_proportions | intersection_count | street_length_total | street_segment_count | street_length_avg | circuity_avg | self_loop_proportion | osmnx_node_id | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 7 | NaN | 11.0 | 11.0 | 2.0 | 1261.861 | 114.714636 | 3.0 | {0: 0, 1: 0, 2: 0, 3: 11} | {0: 0.0, 1: 0.0, 2: 0.0, 3: 1.0} | 11.0 | 1261.861 | 11.0 | 114.714636 | 1.038343 | 0.0 | 337976.0 |
| 8 | NaN | 13.0 | 13.0 | 2.0 | 2126.471 | 163.574692 | 3.0 | {0: 0, 1: 0, 2: 0, 3: 13} | {0: 0.0, 1: 0.0, 2: 0.0, 3: 1.0} | 13.0 | 2126.471 | 13.0 | 163.574692 | 1.030988 | 0.0 | 337979.0 |
| 9 | NaN | 14.0 | 14.0 | 2.0 | 1870.996 | 133.642571 | 3.0 | {0: 0, 1: 0, 2: 0, 3: 14} | {0: 0.0, 1: 0.0, 2: 0.0, 3: 1.0} | 14.0 | 1870.996 | 14.0 | 133.642571 | 1.048630 | 0.0 | 337983.0 |
| 10 | NaN | 14.0 | 14.0 | 2.0 | 1815.929 | 129.709214 | 3.0 | {0: 0, 1: 0, 2: 0, 3: 14} | {0: 0.0, 1: 0.0, 2: 0.0, 3: 1.0} | 14.0 | 1815.929 | 14.0 | 129.709214 | 1.050192 | 0.0 | 337985.0 |
| 11 | NaN | 14.0 | 14.0 | 2.0 | 1870.996 | 133.642571 | 3.0 | {0: 0, 1: 0, 2: 0, 3: 14} | {0: 0.0, 1: 0.0, 2: 0.0, 3: 1.0} | 14.0 | 1870.996 | 14.0 | 133.642571 | 1.048630 | 0.0 | 337986.0 |
# Load Leciester's embeddings
leicester_emb_df = pd.read_csv(this_repo_directory + "/data/leicester-1864_emb-pool_gnnuf_model_v0-5.csv")
leicester_emb_df.head()
| osmnx_node_id | EMB000 | EMB001 | |
|---|---|---|---|
| 0 | 337976 | -0.214178 | 0.073930 |
| 1 | 337979 | -0.112598 | -0.131984 |
| 2 | 337983 | -0.074082 | 0.194142 |
| 3 | 337985 | -0.095356 | 0.193887 |
| 4 | 337986 | -0.074082 | 0.194142 |
fig = px.scatter(
leicester_emb_df,
x="EMB000",
y="EMB001",
hover_data=['osmnx_node_id'],
width=800, height=800
)
fig.update_layout({"plot_bgcolor": "#ffffff"})
fig.update_xaxes(showgrid=True, gridwidth=1, gridcolor='#cccccc', zeroline=True, zerolinewidth=1, zerolinecolor='#cccccc')
fig.update_yaxes(showgrid=True, gridwidth=1, gridcolor='#cccccc', zeroline=True, zerolinewidth=1, zerolinecolor='#cccccc')
fig.show()
for node in leicester_osmnx_graph_prj.nodes:
if len(leicester_emb_df[leicester_emb_df["osmnx_node_id"] == node]["EMB000"].values) == 0:
leicester_osmnx_graph_prj.nodes[node]["EMB000"] = None
leicester_osmnx_graph_prj.nodes[node]["EMB001"] = None
else:
leicester_osmnx_graph_prj.nodes[node]["EMB000"] = float(leicester_emb_df[leicester_emb_df["osmnx_node_id"] == node]["EMB000"].values)
leicester_osmnx_graph_prj.nodes[node]["EMB001"] = float(leicester_emb_df[leicester_emb_df["osmnx_node_id"] == node]["EMB001"].values)
ox.plot_graph(leicester_osmnx_graph_prj, node_color=[
leicester_osmnx_graph_prj.nodes[node]["EMB000"] for node in leicester_osmnx_graph_prj.nodes],
node_size=3,
figsize=(10, 10))
(<Figure size 1000x1000 with 1 Axes>, <Axes: >)
ox.plot_graph(leicester_osmnx_graph_prj, node_color=[
leicester_osmnx_graph_prj.nodes[node]["EMB001"] for node in leicester_osmnx_graph_prj.nodes],
node_size=3,
figsize=(10, 10))
(<Figure size 1000x1000 with 1 Axes>, <Axes: >)
Checking correlations between EMB000 and EMB001 and all the basic statistics from OSMnx. No clear correlation found.
leicester_pairplot_df = pd.merge(
leicester_emb_df[["osmnx_node_id", "EMB000", "EMB001"]],
leicester_osmnx_basic_stats[["osmnx_node_id",
"n", "m", "k_avg", "edge_length_total", "edge_length_avg",
"streets_per_node_avg", "intersection_count", "street_length_total",
"street_segment_count", "street_length_avg", "circuity_avg"]],
on="osmnx_node_id"
).drop(columns=["osmnx_node_id"])
print(leicester_pairplot_df.corr(method="kendall"))
EMB000 EMB001 n m k_avg
EMB000 1.000000 -0.170566 -0.138311 -0.068330 0.376507 \
EMB001 -0.170566 1.000000 -0.226253 -0.212787 0.036726
n -0.138311 -0.226253 1.000000 0.891733 0.013712
m -0.068330 -0.212787 0.891733 1.000000 0.132379
k_avg 0.376507 0.036726 0.013712 0.132379 1.000000
edge_length_total 0.208221 -0.245809 0.589867 0.665304 0.305397
edge_length_avg 0.579544 -0.021938 -0.296360 -0.234006 0.292995
streets_per_node_avg 0.430593 -0.420593 0.097850 0.117131 0.151434
intersection_count -0.019241 -0.302492 0.826980 0.810128 0.057609
street_length_total 0.190095 -0.315328 0.617244 0.656864 0.211502
street_segment_count -0.069979 -0.284915 0.902224 0.878092 0.040801
street_length_avg 0.588756 -0.014573 -0.292572 -0.224905 0.323991
circuity_avg -0.065721 0.225035 -0.118638 -0.118920 -0.000375
edge_length_total edge_length_avg
EMB000 0.208221 0.579544 \
EMB001 -0.245809 -0.021938
n 0.589867 -0.296360
m 0.665304 -0.234006
k_avg 0.305397 0.292995
edge_length_total 1.000000 0.104432
edge_length_avg 0.104432 1.000000
streets_per_node_avg 0.321869 0.340156
intersection_count 0.674507 -0.166574
street_length_total 0.879629 0.080524
street_segment_count 0.635415 -0.232123
street_length_avg 0.108293 0.904994
circuity_avg -0.089911 0.048389
streets_per_node_avg intersection_count
EMB000 0.430593 -0.019241 \
EMB001 -0.420593 -0.302492
n 0.097850 0.826980
m 0.117131 0.810128
k_avg 0.151434 0.057609
edge_length_total 0.321869 0.674507
edge_length_avg 0.340156 -0.166574
streets_per_node_avg 1.000000 0.279544
intersection_count 0.279544 1.000000
street_length_total 0.389084 0.739393
street_segment_count 0.197164 0.910502
street_length_avg 0.316873 -0.170992
circuity_avg -0.153925 -0.143807
street_length_total street_segment_count
EMB000 0.190095 -0.069979 \
EMB001 -0.315328 -0.284915
n 0.617244 0.902224
m 0.656864 0.878092
k_avg 0.211502 0.040801
edge_length_total 0.879629 0.635415
edge_length_avg 0.080524 -0.232123
streets_per_node_avg 0.389084 0.197164
intersection_count 0.739393 0.910502
street_length_total 1.000000 0.687454
street_segment_count 0.687454 1.000000
street_length_avg 0.084041 -0.233779
circuity_avg -0.106058 -0.132164
street_length_avg circuity_avg
EMB000 0.588756 -0.065721
EMB001 -0.014573 0.225035
n -0.292572 -0.118638
m -0.224905 -0.118920
k_avg 0.323991 -0.000375
edge_length_total 0.108293 -0.089911
edge_length_avg 0.904994 0.048389
streets_per_node_avg 0.316873 -0.153925
intersection_count -0.170992 -0.143807
street_length_total 0.084041 -0.106058
street_segment_count -0.233779 -0.132164
street_length_avg 1.000000 0.043635
circuity_avg 0.043635 1.000000
Checking the ego-network of the nodes with the most extreme values in each direction (min and max) for both UMAP0 and UMAP1
leicester_emb_df[leicester_emb_df["osmnx_node_id"] == 192297276]
| osmnx_node_id | EMB000 | EMB001 | |
|---|---|---|---|
| 2657 | 192297276 | 0.899455 | 0.883451 |
ego_192297276 = nx.generators.ego_graph(
leicester_osmnx_graph, 192297276,
radius=max_distance, undirected=True, distance="length")
ego_192297276_prj = ox.project_graph(ego_192297276)
ox.plot_graph(
ego_192297276_prj,
node_size=[20 if node == 192297276 else 5 for node in ego_192297276_prj.nodes],
node_color=["#e41a1c" if node == 192297276 else "#ffffff" for node in ego_192297276_prj.nodes],
figsize=(5, 5)
)
(<Figure size 500x500 with 1 Axes>, <Axes: >)
leicester_emb_patters_df = leicester_emb_df.copy()
# from sklearn.cluster import DBSCAN
# leicester_emb_df_clust = leicester_emb_df[["EMB000", "EMB001"]].dropna()
# clust = DBSCAN(eps=0.015, min_samples=100)
# leicester_emb_patters_df["clust"] = clust.fit_predict(leicester_emb_df_clust)
# leicester_emb_patters_df["clust"] = leicester_emb_patters_df["clust"].apply(lambda x: str(x) if x>-1 else None)
# leicester_emb_patters_df.head()
# leicester_emb_df.head()
# fig = px.scatter(
# leicester_emb_patters_df,
# x="EMB000",
# y="EMB001",
# color="clust",
# hover_data=['osmnx_node_id'],
# width=800, height=800
# )
# fig.update_layout({"plot_bgcolor": "#ffffff"}, width=800, height=800)
# fig.update_xaxes(showgrid=True, gridwidth=1, gridcolor='#cccccc', zeroline=True, zerolinewidth=1, zerolinecolor='#cccccc')
# fig.update_yaxes(showgrid=True, gridwidth=1, gridcolor='#cccccc', zeroline=True, zerolinewidth=1, zerolinecolor='#cccccc')
# fig.show()
def bivariate_colour(x, limits):
if x[0] is None or x[1] is None:
return None
else:
if x[0] <= limits[0, 0]:
if x[1] <= limits[1, 0]:
# return "#e8e8e8"
return "#e8e8e8"
elif x[1] <= limits[1, 1]:
# return "#cbb8d7"
return "#e4acac"
else:
# return "#9972af"
return "#c85a5a"
if x[0] <= limits[0, 1]:
if x[1] <= limits[1, 0]:
# return "#e4d9ac"
return "#b0d5df"
elif x[1] <= limits[1, 1]:
# return "#c8ada0"
return "#ad9ea5"
else:
# return "#976b82"
return "#985356"
else:
if x[1] <= limits[1, 0]:
# return "#c8b35a"
return "#64acbe"
elif x[1] <= limits[1, 1]:
# return "#af8e53"
return "#627f8c"
else:
# return "#804d36"
return "#574249"
leicester_emb_quantiles = leicester_emb_df[["EMB000", "EMB001"]].quantile([1/3, 2/3]).values.transpose()
leicester_emb_patters_df["bivariate_colour"] = leicester_emb_patters_df.apply(
lambda x: bivariate_colour([x["EMB000"], x["EMB001"]], leicester_emb_quantiles), axis=1
)
leicester_emb_patters_df.head()
| osmnx_node_id | EMB000 | EMB001 | bivariate_colour | |
|---|---|---|---|---|
| 0 | 337976 | -0.214178 | 0.073930 | #e4acac |
| 1 | 337979 | -0.112598 | -0.131984 | #e8e8e8 |
| 2 | 337983 | -0.074082 | 0.194142 | #e4acac |
| 3 | 337985 | -0.095356 | 0.193887 | #e4acac |
| 4 | 337986 | -0.074082 | 0.194142 | #e4acac |
plt.figure(figsize=(7,7))
ax = plt.axes()
ax.set_facecolor("white")
plt.scatter(
x=leicester_emb_patters_df.EMB000,
y=leicester_emb_patters_df.EMB001,
c=leicester_emb_patters_df.bivariate_colour,
s=10, edgecolors='black', linewidth=0.1)
plt.xlabel("Embeddings first dimension")
plt.ylabel("Embeddings second dimension")
plt.show()
fig = go.Figure()
fig.add_trace(go.Scatter(
x=leicester_emb_patters_df.EMB000,
y=leicester_emb_patters_df.EMB001,
mode='markers',
marker=dict(color=leicester_emb_patters_df.bivariate_colour)
))
fig.update_layout({"plot_bgcolor": "#ffffff"}, width=800, height=800)
fig.update_xaxes(showgrid=True, gridwidth=1, gridcolor='#cccccc', zeroline=True, zerolinewidth=1, zerolinecolor='#cccccc')
fig.update_yaxes(showgrid=True, gridwidth=1, gridcolor='#cccccc', zeroline=True, zerolinewidth=1, zerolinecolor='#cccccc')
fig.show()
leicester_osmnx_bivariate = leicester_osmnx_graph_prj.copy()
for node in leicester_osmnx_bivariate.nodes:
node_bivariate_colour = leicester_emb_patters_df.loc[leicester_emb_patters_df["osmnx_node_id"] == node]
if node_bivariate_colour.empty:
leicester_osmnx_bivariate.nodes[node]["bivariate_colour"] = "#000000"
leicester_osmnx_bivariate.nodes[node]["node_size"] = 1
else:
leicester_osmnx_bivariate.nodes[node]["bivariate_colour"] = node_bivariate_colour["bivariate_colour"].values[0]
leicester_osmnx_bivariate.nodes[node]["node_size"] = 7
ox.plot_graph(
leicester_osmnx_bivariate,
node_color=[leicester_osmnx_bivariate.nodes[node]["bivariate_colour"] for node in leicester_osmnx_bivariate.nodes],
node_size=3,
figsize=(10, 10))
(<Figure size 1000x1000 with 1 Axes>, <Axes: >)
ox.plot_graph(
leicester_osmnx_bivariate,
node_color=[leicester_osmnx_bivariate.nodes[node]["bivariate_colour"] for node in leicester_osmnx_bivariate.nodes],
node_size=[leicester_osmnx_bivariate.nodes[node]["node_size"] for node in leicester_osmnx_bivariate.nodes],
bgcolor="#ffffff", edge_color="#000000", edge_linewidth=0.1,
figsize=(12, 12))
(<Figure size 1200x1200 with 1 Axes>, <Axes: >)
import geopandas as gpd
leicester_gdf = gpd.GeoDataFrame(
leicester_osmnx_graph_prj_df,
geometry=gpd.points_from_xy(
leicester_osmnx_graph_prj_df.lon,
leicester_osmnx_graph_prj_df.lat
),
crs="EPSG:4326"
).merge(leicester_emb_patters_df, on='osmnx_node_id', how='left').merge(leicester_osmnx_basic_stats, on='osmnx_node_id', how='left')
leicester_gdf.head()
| y | x | street_count | elevation | elevation_aster | elevation_srtm | lon | lat | osmnx_node_id | ref | ... | edge_length_avg | streets_per_node_avg | streets_per_node_counts | streets_per_node_proportions | intersection_count | street_length_total | street_segment_count | street_length_avg | circuity_avg | self_loop_proportion | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 5.829804e+06 | 622151.977595 | 3 | 72.0 | 35 | 72 | -1.196195 | 52.604506 | 194739 | NaN | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 1 | 5.829991e+06 | 622098.041002 | 3 | 72.0 | 45 | 72 | -1.196922 | 52.606196 | 1551014281 | NaN | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 2 | 5.828827e+06 | 622259.813792 | 2 | 79.0 | 57 | 79 | -1.194965 | 52.595696 | 326312 | 21 | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 3 | 5.830107e+06 | 622077.742140 | 3 | 79.0 | 43 | 79 | -1.197179 | 52.607245 | 326320 | 21 | ... | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
| 4 | 5.829673e+06 | 622220.645785 | 3 | 74.0 | 35 | 74 | -1.195230 | 52.603314 | 2627867454 | NaN | ... | 160.60625 | 3.0 | {0: 0, 1: 0, 2: 0, 3: 11} | {0: 0.0, 1: 0.0, 2: 0.0, 3: 1.0} | 11.0 | 1927.275 | 12.0 | 160.60625 | 1.021342 | 0.0 |
5 rows × 30 columns
# leicester_gdf.dropna(subset=['clust']).explore(
# column="clust",
# marker_kwds={"radius": 7}, style_kwds={"stroke": False},
# tiles="Stamen Toner"
# )
leicester_gdf[leicester_gdf["bivariate_colour"]!="#000000"].dropna(subset=["EMB000"]).explore(
# column="EMB000", cmap="YlOrRd",
color="bivariate_colour",
marker_kwds={"radius": 7}, style_kwds={"stroke": False},
legend=True,
tiles="Stamen Toner"
)